Enfoque de Género en la Cultura de las Organizaciones

To what extent can inter-country differences in relative income poverty rates be attributed to differences in tax/transfer systems? Aggregate measures indicate that social expenditure forms a low

proportion of national income in Ireland, much lower than in the EU countries with the lowest relative poverty rates.4 _{Aggregate level} comparisons of “welfare effort” and relative income poverty rates suggest that there is a relationship. Data for 1999 suggest that an extra percentage point on social security as a proportion of GDP is associated, on average, with a reduction of 0.4 percentage points in the proportion of persons falling below 60 per cent of median income.

There is, however, a more direct way to examine the possible impact of differences in tax and welfare structures on inter-country differences in relative income poverty. This involves using a tax- benefit model which can examine the first-round impact of simulating a “foreign country” policy as well as its own domestic policies to arrive at a more precise estimate of how much policy differences contribute to the explanation of differences in poverty rates. Callan et al. (2004) use SWITCH, the ESRI tax benefit model, in conjunction with information on Danish policies generated in the construction of EUROMOD, a tax-benefit model for Europe, to undertake such an analysis. The year for which the comparison was undertaken was 1998.

The impact of welfare benefits on relative income poverty depends crucially on how benefit payment rates relate to the poverty line, which is in turn related to average incomes. For this reason, we focus on benefit payment rates in Ireland and in Denmark in relation to national average earnings. When “importing” the Danish policy into the Irish setting, we ensure that the payment rate provides the same proportion of average earnings as in the original Danish setting.

A further key difference is that a greater proportion of the Danish population is covered for key social insurance schemes than in Ireland. If eligibility depended on contributions then past employment history and contribution record would be critical. But for some of the biggest social insurance schemes in Denmark – including pensions – eligibility is linked to residence, so that how much is paid in pension depends on the length of stay in the country, not on former income or contribution record. In order to capture this difference, we simulate a “Danish-style” system in Ireland under which the payment rates for non-contributory and contributory Old Age Pensions are the same, and scaled to provide the same level of income in relation to average earnings as the Danish pension.

As might be expected, there is a substantial cost associated with moving to Danish-style payment rates and coverage. The net cost (after tax revenue from increased payments) is of the order of €2,400 million per annum. To arrive at a consistent scenario for evaluation of the impact of such a policy, we examine a situation in 4 _{The adjustments suggested by Lawlor and McCarthy (2003) would not alter this}

which the standard and top rates of income tax are raised by 11 percentage points each (i.e., from 24 to 35 per cent, and from 46 to 57 per cent). Clearly, such substantial changes in welfare payments and tax rates would have significant implications for labour market behaviour (see Callan et al. (2003) for estimates of likely behavioural responses to tax/transfer policy changes), and we return to this critical issue below. Nevertheless, it is of interest to explore the potential first-round impact of this change to tax and welfare policies on relative income poverty, before any consequent changes in behaviour.

The simulation results show substantial falls in the Irish income poverty rate at 60 per cent of median income, the application of the Danish structure/support levels reducing the rate by 7 percentage points. There is little or no impact on poverty at the lower cut-offs. This means, loosely speaking, that differences in social protection could account for about two-thirds of the difference in actual relative poverty rates between the two countries with the 60 per cent threshold, which tends to be taken as the “headline” social inclusion indicator at EU level. This simulation takes into account the need to increase taxes, but does not take account of behavioural responses in the labour market.

What are the broader lessons to be drawn from this analysis? Atkinson has pointed out that:

Social investment in improving labour market skills and employability, or an ‘active welfare state’, is an important part of anti-poverty policy, but is not a complete substitute for social spending (Atkinson, 2000). Thus, for anti-poverty policy to make progress requires enhanced education and employment opportunities and improved income supports. Both elements are necessary – neither is sufficient on its own to ensure success in combating relative income poverty.

The success of countries such as Denmark and the Netherlands in keeping relative poverty at low levels over a sustained period depends crucially on both of these factors: a high employment rate and a comprehensive welfare system ensuring that those without income from employment have an adequate income. Each of these factors is necessary, but neither on its own can be regarded as sufficient to keep relative poverty at a low level. Since the mid- 1980s Ireland has made the transition from a labour market with relatively low participation rates and high unemployment to one with high employment and low unemployment. This represents a major achievement, and one of the two key elements identified above as distinguishing countries with low relative poverty rates such as Denmark and the Netherlands from others. Over this period, however, relative income poverty in Ireland has remained higher than the EU average. Comparison with “best practice”, in the EU countries who do best on this indicator, suggests that achieving low rates of relative income poverty risk would require a more comprehensive safety net and higher rates of welfare payment. In considering this issue further it will be necessary to clarify how Scandinavian countries and the Netherlands have managed to

combine high replacement rates with high employment rates, and to be aware of current trends and issues in the management and development of these systems.

4.